SA draft redistribution published

The South Australian state redistribution was released two weeks ago, but with all of the other elections I have only now had time to complete my analysis. This wasn’t helped by the SA Electoral Districts Boundaries Commission (EDBC) not publishing a shapefile. I had hoped to move away from drawing my own boundary files and simply using the official shapefiles, but in this case it was necessary. So you can find the boundary files on my maps page.

This redistribution was relatively modest. No seats have been abolished, created or renamed. 27 out of 47 seats have been left completely untouched. 3.5% of the state’s voters have been moved into a different seat.

The biggest change was in Stuart, with 27.9% of the new seat’s electors previously residing in another seat. This electorate has lost the eastern half of Port Augusta to Giles and expanded south to take in areas closer to Stuart MP Geoff Brock’s old seat of Frome.

Generally some of the biggest changes have been in regional seats to the north of Adelaide, and in the northern suburbs of Adelaide.

The seat of Giles has changed quite significantly, with 24% of voters new to the seat. Giles previously covered Whyalla and half of Port Augusta, but has gained the remainder of Port Augusta and lost its outback territory, to become a much smaller seat geographically.

Stuart had covered Port Augusta at least as far back as 2002, but has now lost it entirely. After gaining Port Pirie in 2022, it has now expanded even further south, although the seat still stretches to the NT border and gained some more outback areas.

The neighbouring seat of Frome has 11.9% new voters, thanks to the seat expanding closer to the northern fringe of Adelaide. The northern Adelaide seats of Elizabeth and King also have about one eighth of their enrolment as new to the seat.

The EDBC is the only commission in the country to calculate its own margins when redistributions are published, which is a relic of the now-repealed fairness clause which required the EDBC to draw boundaries so it would produce a “fair” outcome where a party winning the 2PP would win a majority of seats.

But I prefer to do my own margins, since they have some differences in methodology. For example, they try to take account of population trends, which means that a faster-growing part of a seat will count more for the new margin than a slower-growing area. But I consider that a separate process. I don’t think you can necessarily assume newly-added voters in a fast-growing area will simply replicate the existing votes. So I simply try to show how the last election’s results would look on new boundaries.

The EDBC process can produce some weird results. It means that sometimes they will produce changed margins for seats that haven’t been redrawn at all. For example, Florey and Playford have been entirely untouched but the EDBC estimated the Labor margin would increase by 2.1% and 2.7% respectively. The EDBC also doesn’t attempt to calculate 2CP margins for non-classic seats. Everything is Liberal vs Labor.

For a bit of transparency, there were four seats where my calculations mixed together Labor vs Liberal areas and Independent vs Liberal areas. What I did was take the EDBC’s 2PP estimates per SA1 for the old seats where the final 2CP was Independent vs Liberal and take the difference between 2PP and 2CP and apply that to the inconsistent areas.

Of the 20 seats which had a change to their boundaries, the margin changed by at least 0.1% in eighteen seats, and by 1% or more in six seats. Those seats are:

• King goes from 2.9% to 3.9% ALP
• Heysen goes from 1.9% to 2.8% LIB
• Morphett goes from 4.5 to 5.1% LIB
• Frome cut from 8.1% to 3.3% LIB
• Giles cut from 21% to 17.3% ALP
• Black cut from 2.7% to 1.1% LIB

If you look at the uniform swing needed to deprive Labor of a majority, make the Liberals the biggest party, or give the Liberals a majority, none of these figures have changed. They are, respectively, 4%, 5.6% and 10%.

Okay just to finish up, this map can be toggled to show the old and draft boundaries for each area. Following this map, you can see a table of my margin estimates for each seat compared to the actual 2022 results, and a table of my estimated primary votes.

This is probably the last coverage I’ll give to South Australia until this redistribution process is finalised – for the rest of the week I’ll be back to covering the NSW council elections.

Margin estimates

Primary vote estimates

Liked it? Take a second to support the Tally Room on Patreon!